What's New in Photoshop CS5 & Photoshop CS5 Extended

[Photoshop CS5 and Photoshop CS5 Extended] [Bryan O'Neil Hughes - Senior Product Manager, Photoshop] [no audio] Okay, let's get started. You guys can all hear me extremely well. I might have to whisper. All right. My name is Bryan O'Neil Hughes. I am the senior product manager on the Photoshop team. There are two product managers, two versions of Photoshop--Photoshop and Extended. Zorana Gee, who does Extended, is also doing some courses specific to Extended. The idea here is to give you an overview of CS5 with a little bit of Extended. We're going to cover a lot of information in a very short period of time. I'm not trying to preload your reviews or anything, but we'll start off doing something slightly unorthodox, and I'll give you my email address. The reason for that is we cover a lot of information. You guys are covering all sorts of information in all sorts of courses. You're going to get tons of stuff thrown at you. Later I'll open things up to questions for a few minutes, but you might be that person in the class that doesn't want to raise your hand, or you might just realize later, "What was that reference he was talking about?" So do feel free to send me an email. I get a lot of them. I can't promise I'll respond immediately, but I will get back to you if just to send you some references to what we talked about today. So everybody hopefully has that copied down. It's pretty simple. bhughes@adobe.com Let's get started here. The very first thing that we're going to talk about is something that you don't see, but it's something really, really important, and we're going to drive it through something new that you do see, which is Mini Bridge. We'll talk about that in a second. That's the migration to 64 bit on the Mac. As of this last time around, we are now 64-bit native cross platform. That was a big deal. It was a lot of work. In fact, if we were to go back a couple years, we had already done it for CS4 on Windows. If we were to go back and we started doing this for CS5, we knew we were going to have to rewrite over a million lines of code. We had a tremendous amount of work ahead of us, and we thought, "It's worth it. It's the right thing to do." "We want to make sure that Photoshop is faster." "We want to make sure that it's future proof." "We want to make sure that we can address the growing needs "of people throwing images to Photoshop." And images certainly aren't getting any smaller-- multilayered files, HDR files, video files. So we knew all the reasons we needed to do it, but we were a little concerned that CS5 would be a whole lot more than just 64 bit. Well, I'm really excited that we get to talk about this up front, that I get to explain it, and that I get to show you all the things that we put in there because I can tell you, Photoshop has been around for 20 years, I've been on the team for 11 years; this is easily our most innovative release to date. It's certainly the one that's leaned the most on other people, and you're going to see a lot of magic today, and, indirectly, 64 bit is to credit for that because we kind of freaked out and we said, "We're going to need all the help we can get from throughout the company "and even beyond." So just bottom line on 64 bit, what it means is that if you have a lot of memory you can address it in Photoshop now. A 32-bit application can address just under 4 gigabytes of RAM. A 64-bit application can address--it's a theoretical limit. There aren't any computers that give you enough RAM for Photoshop to soak it all up. More is more. And what it means is 10 to 12 percent speed-up in day-to-day operations. In the event that you're throwing a really large file at it, it could be up to 15 times faster. So multilayered files, hundreds of layers. We see people with thousands of layered files--again, video, HDR, 3D, things like that. They love RAM. You've probably seen where Photoshop or other applications everything is going fine, and then you get the beach ball and things just drag to a halt. And that's because you're paging out to disk. You've run out of RAM, and you're writing to disk and you're reading it back. In a world where you have more RAM and you can take advantage of it, that's no longer a problem. All right. So that's a little bit on 64 bit. The next thing I want to talk to you about is Camera RAW. I come from a photographic background. I'm used to talking to a lot of photographers, most of which are shooting RAW now, but it's important to say right up front everything that I'm going to show you in Camera RAW can be done to JPEGs and TIFFs as well. In fact, if you're newer to Photoshop or you're just getting into Photoshop, this is a great inroad to learning it because it's very friendly, it's nondestructive, easy to use. Camera RAW is in its sixth revision. It's the same engine that powers Lightroom. I don't know how much you'll see of what I want to show you onscreen here because it's a little darker. We can read over 300 proprietary RAW formats, but again, because we can write JPEGs and TIFFs, you could throw an image from your phone through here, you could throw an image through your DSLR. We've rewritten the entire RAW processing engine this time around. We did it for a few different reasons. We wanted to give you sharper images, better color fidelity, some new special effects that I'll talk about, but the most important thing and hopefully one we can see onscreen here-- I know we'll be able to see it a little bit--is noise reduction because going back to either cell phone images or high end DSLRs, noise is a big problem. If you have a tiny little lens, you're not getting a whole lot of information, and that generates noise. And if you have a high end DSLR, they shoot crazy high ISOs, and that means noise. So historically, on an image like this if we wanted to remove the noise, for one thing, I should be real honest. I don't think we've done the best noise reduction that we could in the past. But even with the best out there, what you'd do is you'd get rid of the noise but you'd lose all that color glow and you'd lose the detail as well. Let me show you what we can do now. The first thing we're going to do--and I wish you could all step behind me for this part-- we're going to take out the color noise, which is in the black area. You probably didn't even see very much happen, but believe me, that black area there is heavily polluted with color noise. You've seen that before. Trust me when I tell you that's gone. But the more important part is the luminance noise; it's that crunchy noise that's all around the rest of the image. And not only can we smooth that out, but by sliding that Detail slider we can preserve all of the detail in there. This is really important because we're removing all of that noise, but we're preserving all of the detail, including the glow here. And this is a problem anyone and everyone has. It's really easy to solve it. What we're doing uniquely is preserving all of the fidelity of the image there. Let me back up a little and show you something else we're doing. We're going to show you this in Photoshop as well. Another thing we can do is lens correction. Lens correction historically has been a very manual process. This is another thing that afflicts camera phones all the way up to DSLRs. Every lens, no matter how good it is, has some problems. It has some distortion, it has what's called vignetting-- the natural fall-off in the corners--and it has chromatic aberration, which is that color fringing. You'll often see you have a picture with phone lines going across the sky. That's always been a manual process to remove those things. We can automatically do that now because we know which camera you're shooting with, we know which lens, and we've built a variety of different profiles for them. As of today, I think it's one of about a million things we announced today. It might not hit the wire until later. We have a little standalone application that will show you all the user-generated profiles as well. So if you're in here and you've got a camera and lens and you don't see a profile for yours, you can search, and if someone else has created one, you can pull that down. And we've posted a little Lens Profile Creator as well. So that's automating what has been a very manual process. I want to show you something that for anyone compositing with Photoshop, I think this is a really important thing to know. I think I mentioned that Camera RAW is also the engine of the Develop module in Lightroom. And when I talk to a Lightroom user or a photographic user about additive grain, for one thing, it sounds like I'm talking out of both sides of my mouth because I just talked about removing noise and now I'm talking about adding grain, but for another thing, they think of the aesthetic of film and changing the look of an image. And we can certainly do that. We can now add grain, and we can make this look like T-MAX 3200, to show my film roots. But the reason that I think grain is important to you guys and to a lot of people going into Photoshop is let's say I have an image like this that I took in Africa a few years ago. It was shot in the morning, ISO 100 or 200, and then I've got some images of sunset later in the day shot at 1600, 3200. I put those in a book next to each other, and even if I process them the same way, the grain structure is different, and they don't quite look right side by side. If I want to composite something from one end to the other, forget it. It's going to look really strange. You can tell there's something seriously wrong in those two images living together. By adding just a taste of grain you can establish a consistency in the grain structure of the images. If you know you're going to be compositing images and they come from different sources, this is a great way to make them look that much better. We'll talk about selections and stuff like that in a second. But that's a really good one to know about. Just to give you an idea, those of-- Just out of curiosity, how many of you guys are shooting RAW or working with RAW files? Okay. Wow, that's gone way up in the last couple years. That's great to hear. Let me give you a quote that I will paraphrase from Tyler Stableford. He's a Canon Explorer of Light, a distinguished photographer, he took this cover image, and he's been saying what a lot of people are saying about that new RAW processing engine. He refers to it as $1000 worth of glass. He said it's like putting a brand new L Series lens in his case on his old images. It's that much sharper, it's that much less noise, it's that much better color fidelity. So one of the things I encourage you to do for those of you working with RAW files is revisit those RAW files in the new version of Camera RAW because that new process engine just yields much, much better results, especially with those high ISO images. All right, moving right along. Let's actually jump into Photoshop here. The first thing we're going to talk about is HDR. We're in Mini Bridge. I just sort of touched on that ever so briefly earlier. The whole idea with Mini Bridge is--back to Photoshop 7, it was a really exciting time. We had the File Browser, we had the Camera RAW plug-in coming out, this mass proliferation of digital cameras, and we realized that Photoshop users were going from one image at a time to dozens, hundreds, thousands. The Suite came out, they've got all these different sorts of images, so we came out with the File Browser which ran from within Photoshop, let you see all these different assets. But it wasn't quite fast enough, so we made it its own standalone application which became Bridge. The only problem with Bridge is you had to know to launch Photoshop and to launch Bridge and to go back and forth between the two of them. Some people were perfectly fine with that, other people-- and as the Photoshop product manager, I like these people a lot-- just want to stay in Photoshop; they don't want to leave. And what they want is that File Browser experience with speed. So that's what Mini Bridge is. Mini Bridge is just calling Big Bridge, as we call it, in the background there and it's loading it in a panel. And like any panel, I can change the size of that, I can lay it out as a filmstrip, I can put it on a separate monitor, can see every different asset that Photoshop can see. I like to remind people that from the Hubble Space Telescope to the phone in your pocket, Photoshop can see just about any imagery out there. So in this case we're actually going to trigger an operation from here as well. It's a new feature called Merge to HDR Pro. HDR stands for high dynamic range, also called 32 bit. Essentially, the idea here is you've got these digital cameras that can take all of these pictures. In addition to just taking a lot of pictures, what do you do with all that information? What do you do with similar shots? In this case we're going to extend the dynamic range of this imagery. Your eye can see a lot more than a camera. If I take a picture of a person standing in front of a bright window, I can either expose for the background and get a nice view out the window and the person will be a silhouette, or I can expose for the person and the background will be blown out. Your eye can do much, much better than that. And it turns out if you feed your camera or if you feed Photoshop three, five, in this case eight--we've gone way beyond--images, you can merge them together and you can get a similar aesthetic. There are two ways to go with this. You can go really wild, crazy. HDR will do that. Or you can do photorealistic, which is what we're trying to yield right away here. We've made it really, really easy to trigger these operations. You just select the images you want, throw it at HDR Pro, you get this really friendly dialog, it shows the images we have across the bottom. We have what's called a tone mapped result there, which is to say it's taken all of the overexposed and underexposed images and put them into one. We've got a variety of different presets here. They're user configurable as well as the ones that we ship with. Again, I'll hearken back to my film roots and say that this one looks a little bit like Fuji Velvia hypersaturated. Everything is looking fine here except we've got this problem, these ghosts here. And this is a problem with anyone who is trying to do HDR. We know quite a bit about aligning layers and files, we know quite a bit about building panoramas. And in this case I have successfully aligned eight images, but the problem is the barbed wire was blowing between them. Even if all the content is aligned, that's blowing between them, trees, sky, water, whatever it might be is moving, and the very things you'd want to take HDR images of, you haven't been able to in the past because of these artifacts. And what you see people doing is really exaggerating their HDR conversions. I don't think they do it so much because they love the look, although we'll look at that aesthetic in a second, I think they do it to camouflage these artifacts. I think a lot of people would like to just yield a much better-looking image but this gets in the way. This is one of those things where for as many talented people as we have at Adobe and on the Photoshop team, we needed to reach beyond. And there was a gentleman at Dolby Labs that had some technology that he had never put into a product, and we had him come work for us for a while just to solve this problem, and he put in this tiny little checkbox here called Remove Ghosts. And if I click on that--take my hands off the keyboard, and watch the screen-- it's going to look through every image there. And watch what happens to the barbed wire automatically. Thank you up front there in the red sweater. I appreciate it. [audience laughter] That's a big deal for HDR. That's something that you really haven't been able to do in that way at all. It means you can do HDR captures of water, things blowing around, all the barbed wire you ever dreamed of. And down here you see that it's shown us which image it locked on to. It mapped the content of that image. A lot of you guys are doing stuff for the Web, so I want to show you what happens when this stumbles just so you know what to look for. So we're going to feed it four JPEGs. Ideally, you'd be throwing this RAW files, but I know that's not necessarily the way it works, and I should tell you I'm breaking all the rules with these. I'm shooting off tripod, my exposures are all over the place. I try to use the software the way I think people will use it. I try to shoot as many of my own demo files as possible. And of course I have a preset for this as well, Boats. And there's that--four images. Remove Ghosts. Because these are JPEGs, it's going to lock on to the image it thinks is correct, but you'll notice there's a bunch of posterization up there. Well, not to fear. Not only can we override that, but I actually like the image that had the bird in it better with the guy's reflection in the water. So I just click on the image I want, and it's going to switch the mapped content to that. Now I get my bird back, and I've got this guy reflected in the water. So even when it fails you can override it. You have quite a bit of control. And I've thrown really, really nice RAW files at this and really poor quality JPEGs, So this is three to as many images as you want to use. A lot of people ask what's the sweet spot. Three to five images ideally stabilized, but none of these are stabilized. You certainly don't have to do that. Okay. So that's high dynamic range. That's if you have multiple images. What if you don't? What if you want the crazy aesthetic? Yes, we can do things like this. Just so you know, that can be done. Some people love this, some people are nauseated by it. It's very possible. I don't show it as part of my demo yet, but it's very possible. So what if you want the HDR aesthetic but you don't have multiple images? Maybe you've got an image off a phone, maybe you've got an older image, maybe the person who handed you the asset says, "Make it look like HDR." So we have a feature, and this is pretty neat. I was meeting with Scott Kelby of the NAPP, the National Association of Photoshop Professionals. I had shown him all this stuff and he said, "It would be really cool "if I could do everything you just showed me to a single image." And I sent a mail to an engineer. I was at Photoshop World before this came out. By the time I got off the plane, the engineer had what I'm about to show you up and running and it just blew my mind. Easily the coolest thing about my job is that sort of interaction with our people back home. So this is what we can do. That's the image as I took it on the left, and there it is run through HDR Toning, as we call it, faux HDR, replicating that aesthetic, really crunchy, glossy. You see this all the time, especially in automotive advertising. And you can get a look that you couldn't get otherwise really easy. You can also get some interesting creative effects. I'll show you how this works. And again, this could be any sort of file. Up here, Image Adjustments, and HDR Toning. Because it's tone mapping it, you remember I said that it looks at the shadows and the highlights. Right away the image looks quite different just by virtue of coming in here. There are a couple things we can do. I refer to the Detail slider as the drama slider. If I move this to the right, it gets really crunchy and contrasty. You see this a lot in portraits now. If I move it to the left, it gets really soft. If I do a mixture of vibrance and desaturation, I get this nice ethereal kind of foggy look to it. This is also really handy for skin tones. You can soften skin by taking Detail in the other direction there. You can do all sorts of really interesting things here. We have presets for this as well--single image presets, even high contrast black and white, all sorts of different things. A lot of people think of HDR and they think of color, but it's really good for black and white as well. So that's just a tiny bit there. That's actually become an even more popular feature than normal HDR. I spoke about lens correction really briefly earlier, but this is the sort of image you'd normally get where you'd really need it. In the past, really, really manual process. I just want to show you how easy this is in Photoshop. I come into Lens Correction--again, my hands are off the wheel here. If I toggle Preview, that's what we came in with. And that's what we came up with by doing nothing. Notice it's a Nikon D3 16mm lens, 2.8, shot at f/16. We've built a bunch of profiles, we keep building them, we're releasing them all the time just like the RAW profiles. And if you're hooked up to the Web, say you don't like this particular profile or say you're looking for one you don't have. You can search for that right here. Let's say you have an image that someone gave you or that you pulled off of Facebook or someone sent you in an email and it doesn't have EXIF data, it doesn't have information that tells you what it was shot with. Facebook strips all that stuff away. So in that case let's say I know my friend shoots only all of their pictures with the iPhone. I can come up here, Apple, iPhone 3G, and I'd get a profile for that. So we've got profiles literally for everything there. I want to run ahead in our story for just a second here and tell you a little about what's called JDI or just do it. This is something we borrowed from the After Effects team. I mentioned we were doing all this major work, and in addition to all the big stuff that we were doing, we wanted to focus on some little things as well. It was 20 years on, we're really in touch with our community, we've got some great seminars and workshops and things like this where I get to talk to all sorts of people. We had a long list going of little things we wanted to work on, and we knew we were going to take a week out and just do some of those. So we thought, "In the case of lens correction, "I think a lot more people are going to be using this." "Let's take this as an opportunity to take any of those little JDI features, "those little things, and address them while we're working on this." So I think there are seven of them in here. The first one is the location. It used to live under the Distort submenu, which, if you think about it, is a little backwards. You're undistorting it; you're not distorting it. So it's right there, waterfront real estate there where you can't miss it under the Filter menu. There's a new slider for chromatic aberration to give you more control. Down at the bottom here the controls for the grid, the size of the grid, and the color didn't used to be persistent, which is to say if you were the sort of person using this all the time--and a lot of people do--they'd come in here and they'd change those things every single time they came in. Not the worst thing in the world, but it would be kind of like getting into your car and having to move the seat and the mirrors and all that every single time you got in. It gets kind of old. If we can save them 30, 40 seconds every time they're coming in there and they use it a lot, they're going to appreciate that. So this is just one place where we've done a whole lot of that. So now we get into two things. We get into gratuitous baby shots and also a little more about very fundamental Photoshop functionality, which is selections and masking. What I wanted to come up with here was what I think would be a really representative image for how difficult it is to make selections of hair. So I picked my extremely furry cat and my little boy there. Just to take things even further, I thought, "Okay. We'll do it with my phone." At that point it was an iPhone 3G one megapixel image. So it's a one megapixel JPEG image of a cat. Nightmare to mask that out, right? We can all agree on that. On the left, that's the best we could do in CS4. Again, tough image to mask. We've got some great selection tools, definitely a core competency of Photoshop, but hair has always been really difficult. So let's see what we can do in CS5. We're going to load the same selection, not because it's a great selection--it really isn't. In fact, if you look, the cat's whiskers are hanging out, some of the fur is in, some of the fur is out. But I wanted to use the exact same selection. What we're going to do is we're going to pull up Refine Edge. With the Selection tool, you have a button for it right here. Otherwise it's available in the Select menu. Really, really powerful. We hired a new designer for Photoshop this last time around, a younger guy, fresh set of eyes. As soon as he saw the Refine Edge dialog, he said, "This is a really powerful feature, "but we've got some stuff we can fix here." So one of the things we do is we have all these different ways of viewing this, and we use your image to show which is which. We used to have kind of this little symbol across the top. You guys have all seen it. It looks like abstract art. People didn't know if it was a button or just placeholder or something colorful. It turns out that triggers different views. Now your image triggers those different views, and there are more of them than ever before. We're just going to do three things here. We're going to give this a wider radius, 15. And immediately in this case--I don't know if you can see that back there, but we've picked up individual strands of hair. So right away we are way, way ahead of where we were before. Next we're going to click Smart Radius, and Photoshop goes all around the edge, and it can tell the difference between the soft edge of the fur and the hard edge of the cat's ears. We're going to do just one last thing. We've got this new brush here, and I can come through and I can option click to say, "Go ahead and pull those areas out." So I'm just cleaning up the mask. What's neat about this is it's learning about the mask as I go. So the mask is getting better every single time. Remember that whisker that was hanging out there? I'm just going to draw a line there and it pops right back in. So I've spent a few seconds--less than a minute--on this mask. I used to say when I first started showing this, "That would take me hours before in CS4." I could not have done that in hours. I'd be doing single strands of hair. You'd have to lock me in a prison cell for a month or so. Again, we just give it a wide radius, smart radius, and we touched up a couple pieces of the mask. And that's on a really, really low res image. So that holy grail of selections, we've really made huge strides there. I'm really, really happy with what we've done. We did more user research this last time around than we ever have before. We spent a tremendous amount of time in the field polishing things like this, giving it a better underlying technology. But we really wanted to save people time as well, and I'm going to talk about that a lot today. There's 64 bit, there's GPU, there's all this hardware accelerated performance, and then there's workflow performance. And for a lot of guys like yourselves who are in a production environment, it's those things that really give you back your time. So one of the things we do down here is we say, "What would you like to do next?" "Do you want to output that to a Selection, to a Layer Mask, "to a New Layer, to a New Document, to a New Document With a Layer Mask?" We saw people were doing a lot of these things as soon as they left and we thought, "Let's just give them that option to do that right out of here." "And further, let's remember those settings." So the next time you come through here, people who are doing stuff for catalog or Web, they're doing the same thing over and over and over again all day long, and we really wanted to speed that up. So that's the idea there. If you guys come to my Hidden Gems session later, I'll dive even deeper into some of the other changes we've done. Unfortunately, because of time I have to keep moving, moving, moving here. But there's a lot more to talk about. I'm going to show you two things in this next feature, two different sample files, because it works differently in both cases. This is largely an unknown feature. It came in really late. Really, really powerful. Let me make this brighter so that you guys can see it. There was a time not that long ago that you would never use Brightness/Contrast in Photoshop, but it works really well now as of CS3, so that's an okay thing to do. What I want to talk about here is selective sharpening, just sharpening in this case around the elephant's eyes and the wrinkles here. People do this all the time to give emphasis to their photos. The way that it works right now is people take their image in, they'll duplicate the layer, they'll sharpen the whole thing, and then they'll erase all of the areas that they don't want sharpened. It certainly works, but it's not really that elegant of a solution. What you want is a pressure-sensitive, brush-based tool that you can just sharpen a little bit of it. We have that tool. Going back to our 20 years, we've probably had what I'm about to show you since day one. It hasn't worked extremely well. It does something like this. So if you want artifacts, it's really good for that. If you want rainbow artifacts on your elephant, we've got you covered. But a lot of people don't use that for sharpening, right? So by default there's this new thing called Protect Detail. I'll tell you guys, there's a lot of sharpening technology out in the world and there's a lot at Adobe. This is the best sharpening algorithm you will ever encounter. It's extremely powerful, and the coolest thing about it is no matter how much I apply it, I'm not going to generate any artifacts. So I'm laying it on real thick here, sharpening this elephant like crazy. Give that a second. We can revert that to where we started. Again, I brightened it, so that's why you're seeing that shift. I sharpened him aggressively, and I didn't bring any artifacts into it. It's impressive there. It's great for that use case I described. The thing that I've found it's even more impressive with is things that are a little bit out of focus, that you just kind of gave up on in the past. You didn't even try to sharpen them. So watch what happens here. We're just going to go through this image, sharpen that, sharpen that, sharpen that. I'm laying it on thick, and it's doing a really nice job. These are dramatically out of focus, and we're just making this stuff pop that much more. So it's really great for when you want to go in and just make an area pop or bring it back to focus. Largely unknown. It's the old Sharpen tool. You'll have to retrain yourself to use that. But you don't have to do anything. It just works like that by default. So I encourage you to try that out. So next up we start moving into magic here, and that's probably the most fun to demonstrate is these things that Photoshop does that you just don't understand how on earth does it do that. A lot of people have heard of Content-Aware Fill. We're certainly going to get into that and talk about that. A lot of people don't realize that Content-Aware Fill is actually a property of the Spot Healing Brush as well. The Spot Healing Brush is a very recognizable tool in Photoshop, really powerful tool. There's not a retoucher out there who doesn't use it. But there are a couple of shortcomings with it, and one of them would be something like this. Sorry. Caffeinated demo here. If I were to remove something like that from an image, you've seen this before, right? It gets blurry, it gets soft. Healing Brush doesn't deal very well with lines like that. Another thing it does poorly, the edge of an image. It'll pull in weird content. There's just certain things it doesn't do as well as others. You see there's a little Content-Aware option here. And again, same excuse. If I come over here and messily draw a line here, the difference with Content-Aware is that it looks all around and it knows the texture and the tone and the noise and the grain. That right there, if you're a retoucher or if you're someone who deals with that, that's saving you hours. It really is. This is kind of reserved for the Hidden Gems session later, but I'm going to show it to you anyway because I think it's important. The trick to this if you really want to know how to do this well is-- People ask all the time, "Photoshop is 20 years old. You've got all these features." "Why don't you take some of those old ones out?" And whenever we think about doing that, along comes a demo like this, and you realize every feature has its place. In this case Paths, super valuable in this particular workflow. You drop a path on the line, you give yourself a really small brush, you don't worry about the caffeine problem I talked about before, and you come over here and you stroke the path with the brush, and it takes care of it for you, does it really precisely. It's really quick and easy. You just drop a couple paths on and you stroke those paths. And if it's a straight line, you don't even have to do that. You just click, shift click, and it's gone. And then I'd just delete that path. So that's really the way to do that. So let's take a closer look at Content-Aware Fill. How many of you guys have seen the YouTube video we did for that? Okay. How many of you are aware of Content-Aware Fill, have heard of Content-Aware Fill? Awesome. Great. I'm going to show you a few different things here. Essentially, it's a lot like what I just showed you, but it's the ability to not just do it brushing; it's the ability to fill in regions of an image. There's certain things you can do with this that you just can't do with any other tool. So in this particular case lens flare is a huge problem. It's a nightmare for a retoucher. It takes a lot of work to get something like that out. You can just select that area, delete it, and it'll take care of that-- the texture, the tone, the color. Oftentimes something that crosses a shadow like this-- I might have pulled that selection in too close--really tricky. It just fills in the shadow, the grass really, really nicely. All I'm doing is I'm making a selection and I'm hitting Delete. Okay. So that's little cleanup stuff where it tends to be the most impressive. I'm going to load this selection just out of the interest of speed more than anything. Just about all the files I've shown you have been my own. I was out on an early press tour. We were showing this technology to different magazines and publications, and I had this file provided by Adobe, and I didn't know what the file was for, I didn't know which feature it supported. I tried HDR and I tried Selections, and somewhere along there I thought, "Wow, I wonder if this is for Content-Aware Fill." "I wonder if this is designed for removing something from the image." "That would be crazy if I could take this guy out of the image and put a wall there." And sure enough, that's exactly what this is for. I select the area I want, I click OK, and he just kind of magically disappears from the image. How did we do that? What's neat about it, we didn't take a man-sized piece of the wall and put it there. If you look at it, we actually extend from the edges in there and we've built a wall, we've synthesized a wall where he was standing. So what we've done is we know where he was, we know what the edge of the selection looks like, we know what the rest of the image looks like, and so we're building it in based upon that. What makes it so convincing is again, we know the texture and the tone and the color and all of the properties of that, and so we're able to assemble that in a very, very convincing way there. Someone always asks, "Could you take out the ivy?" Yes, we could take out the ivy. We could do that as well. Let me show you some other examples of this. Really neat, great for retouching, great for a photo just interpreted slightly wrong-- maybe there's a pole growing out of the back of someone's head. There's a lot of things that you can do here just to show you that-- Here's a picture of me. A bird, something like that, really, really easy. A little trickier when you have a picture with a reflection across sand and rocks and more sand and water, but you can remove those things really, really easily. We'll take me out just to be fair. I should mention if you guys play around with this, I'll give you a couple things to know about for when it doesn't work because occasionally it doesn't. That was not one of those times. It did a great job. So let's talk about when it works and when it doesn't. I was really inspired by Katrin Eismann. She does great demos. I had been playing around with this a lot. This is proof that these are in fact my photos. That's me and my little boy again. What I wanted to do is see how other people were using it. And Katrin was using it for these amazing things. She was taking off people's badges. We're all wearing these silly badges. If you see Martha Stewart later and get your picture taken with her, you probably don't want to be wearing that badge. Content-Aware Fill can take that right off, stitch your sweater where it was. So she was doing all this cool stuff with it and I thought, "Well, I'll try some of those things out." So I thought, "I'll just take an image like this and I'll start pulling pieces out of it." So maybe I'll take that bucket out of it, and maybe I'll try to take this raft and maybe get the reflection out of the water. And then when I got to this point I thought, "There's no way." "There's too much happening here in the middle of the image." "There's water and there's a bucket and all of that." It does a really nice job there. Now let's talk about when it doesn't work. Every tenth time I say this it actually does work, so let's hope it doesn't work for the sake of what I'm about to tell you. You see that little bit of grass that we put up there or whatever that was? I'll give it an even better, bigger selection so you can see what I'm talking about. Okay. It didn't work. How dare it? It's really falling down on the job there. So we have to think a little bit like Content-Aware Fill to understand when it doesn't work because you're going to use this--maybe you already have-- and there will be times when it doesn't quite work. If we have a large selection like this, it's looking around the rest of the image, but it only has so much to look at, and that's a really big area. If you give it a smaller area, not only does it have less to fill but it has more to look at. So in those cases what you want to do is just give it little pieces of the image to look at and just go through it in a few different pieces. And you can fill just about anything by doing this if you have patience. And when it seems like it's taking a while, just remind yourself how long it would have taken before. [chuckles] All right. So here's where we started. Pretty remarkable, right? I'm driving this from a trackpad with the Magic Lasso. So last Content-Aware Fill image I'll show you right now, and this one just blew my mind. All the guys that worked on this--and several people worked on this feature-- I showed this to them and they were like, "Wow. I didn't think it was going to do that." It's a five-image pano that I did in Death Valley a few years ago, really liked it. I've barely retouched it at all. It really did look like that. It was a crazy day. It was snowing in one part, raining in another part, 90 degrees in another part. There was every single thing happening within the window of a couple of hours. But if I wanted to do anything with it, I'd have to crop this image like so. You'll notice we now have a GPU overlay on the grid there, little thirds. It's nice, but I'm going to lose 40, 45 percent of my image there by cropping that. I can't find a frame that's shaped like that, so I'm kind of out of luck. And I thought, "Well, I wonder what Content-Aware Fill would do "with that missing region that's there." Let's try. We just select the white area there, hit Delete, take a much-needed sip of my water, and watch our progress here. All right. Pretty cool. I mean, how long would that take? [chuckles] Where would you even start? And there's mystery in this image. Why does the hill go down on the right? Why does it go up on the left? What weather is coming over there on the sky? It's pretty cool. Even the areas where it doesn't have a lot, yeah, it's going to repeat some content, but it gets you so, so incredibly far along. Again, if you want to see some other really interesting edge case uses for this, come join me later for Hidden Gems. I'm going to do that today and tomorrow. But there's all sorts of really cool stuff that you can do with this. That's becoming one of the key uses for photographers. The other thing you'll see a lot, if you straighten something, if you rotate an image, you'll end up with a white area around it, and they have to crop in on it. Depending on the image, you could use Content-Aware Fill to fill those areas. So it's really handy. I think that's why it's been so popular is so many people use Photoshop for so many different things, and this is really easy to use, so it's really a nice blend of a whole bunch of stuff. The next thing I'm going to talk about, I'm not going to spend a tremendous amount of time on it, not because it's not important--it's incredibly important--two reasons really. One, because it deserves a really long discussion about all the things that you can do with it. The other is I would be doing it a disservice because I am not a skilled painter at all. So I'll stumble through it a little bit. I'll show you what we can do. We'll lean on my adorable little boy to help me through it. He makes up for a lot of my demo follies. Let's talk about what's going on here. Paint has been in Photoshop for a long time, and there a lot of people using Photoshop for illustration and painting. Bert Monroy does these amazing things. I don't know if you guys have ever seen Bert's work. He goes from nothing to these photorealistic paintings using Photoshop that are just absolutely astonishing. There's more seminars on this stuff, there's more books on Paint in Photoshop than ever before, but until CS5 we hadn't touched our Paint feature since I think version 7 or so. It had been a long time. And so we figured it was time to take some of the magic we've got and apply it to Paint. So I've got this entirely new brush called the Mixer Brush. If I had a tablet, which I should have sitting right next to me here, this little guy on the left, which is GPU accelerated, when I tilt my stylus it would tilt. And what's cool is if it was the 6D pen that Wacom makes, if I were to rotate the stylus, the tip of the brush would actually rotate. So a fan-shaped brush that actually rotates in my fingertips. All super pressure sensitive. It's got fluid dynamics. I can load and clear paint. I can do all sorts of amazing things. But the most important thing is I can actually mix color. In the past if I were to take green paint and put it over blue paint or green pixels and put them over blue pixels, I would just be doing that. I would just be covering one thing with another. They would have zero relationship with each other. Now if I come over here, I'm actually mixing blue and green, right? If I bring red over to yellow, yellow over to red, I'm actually creating orange. So because I'm mixing all this stuff together, I really have a tremendous amount of power with what I can do. For a guy like me, a hack with this sort of stuff-- Here's Miles. We'll lean on his Paul Newman eyes there. What we can do is we can paint directly over a photo. So you can come in here and you can have a photograph, and you can start mixing your painting and your photo. You've got all these different presets, all of this control, and for someone who doesn't have a background in painting--obviously-- it's a great place to start. For someone who is using really low res files, this is a great way to camouflage those artifacts. For someone who wants to just make something look a lot different or soften it, that's a great way to go. I'll show you a couple of really cool things here. I can drag and drop directly out of Mini Bridge, in this case to Acrobat, just to show you this really quickly. Julieanne Kost, who in addition to being an amazing evangelist and doing these incredible things, was voted one of Fast Company's Most Creative People in the World last year and I can certainly understand why. She can take a brand new feature and just start whipping up stuff like this in a few minutes. Crazy. What she put together here is intended to just show you a 30,000-foot view how powerful this Brush engine is, that you can load it with different properties, you can make it wet or dry, you've got all of these different tips in this new Mixer Brush. I mentioned that you could turn the tip. Imagine the flat brush turned would be a skinnier brush. You can adjust the count, the thickness, the stiffness. You really have a tremendous amount of control. And again, this is showing fluid dynamics--dry, wet, loaded, unloaded, different presets--and Julieanne just scribbling around and having fun. If you're interested in this sort of thing, it's well worth digging deeper, and I'll give you guys some more information at the end of how you can find out more. But it's just remarkable the things you can do with that. Just to show you one last example, one of the guys on the team took this photo and really oversaturated it to the point that we created artifacts. Going back to that thing I was saying earlier, sometimes we leave an old feature in and someone will find a use for it we never would have thought of. In this case a pretty destructive method of color correction created artifacts, but you can then take those artifacts and turn them into paint strokes. So you never know where an old feature is going to come in handy. That's a great example of that. All right. You notice I'm just sort of going backwards and forwards through Mini Bridge. More magic. Definitely this is as magical as it gets here. Back to the safari pictures. We've got this image of an elephant, which came from this. Really easy to extract an image, quick Select and Refine Edge. I did this in a few seconds. Really, really easy. And for the sake of this demonstration--not because this is a problem that afflicts all of you on a daily basis--we want to make it look like this elephant is eating. So if I were to give this to 10 of the most talented retouchers out there, they'd have 10 different ways of doing this. They'd all take quite a little while. They'd probably have different results. I'm sure they'd all be convincing in the end. They'd probably involve a whole lot of layers, a whole lot of subtle transforms, and a ton of blending because there really aren't ways to directly manipulate an image in Photoshop. You can liquefy it, you can push the pixels around, but if you want to just distort something, that's actually quite tricky. We've got this new feature called Puppet Warp. What we're going to do is we're going to come in here, and the great thing about this is not only is it extremely powerful but it's really easy to use. I'm just going to drop three pins on this elephant, and it's going to tell it how it moves. And so I pick up his trunk there, and I can make that shorter, longer, pull that back. Again, probably not a problem that all of you in the room have on a daily basis, but imagine someone is not quite smiling or imagine someone's eye is looking slightly off camera, or someone's hair is blowing out. Whatever might be happening, what you can do is you can take one little piece and you can directly manipulate it. It could be a piece of rope that you turn into cursive writing, it could be a rose stem that you spell something out. There are so many different things that you can do here. Let me give you another real-world example. So in this particular case I could use a picture I took of my friend. We were out taking pictures, and the camera is a little askew. Obviously, a lot of vignetting, fall-off in the corners, and a little bit of distortion. All of those trees are leaning in there. If I ran that through Lens Correction, the trees on the right would stand up. They'd be straight. But the one on the left would still be leaning in. That's a common problem. Unless you're shooting straight on, even Lens Correction can't know the exact perspective that you shot it. So what we want to do is stand up just the one tree on the left. That's pretty tricky. What are you going to do? You're going to select it and copy it and transform it and hope that that fits in. What we could do with Puppet is we could just make a selection of the thing we want-- in this case the tree--and we could come in here and just drop-- You can drop as many pins as you want. The more you drop, the more control you'll have. But in this case we just need three, and we're just going to stand the tree up. We can make it taller, make it shorter. We've got all sorts of control over that. These can be smart objects, they can go behind other layers or in front of them, they can go behind and in front. All sorts of things that you can do here. The engineer that implemented this, every other day it felt like he was coming by my office saying, "Should I put this in? Should I put smart object support in?" "Should I put this or that?" And my answer to all of it was, "Yes, yes. More, more." No matter what you think, people will take it further than we imagined. And we certainly know that of our users. So the only thing that's more fun than that is actually applying Puppet Warp to a person. I'll make Nicole a little smaller there. Same idea. Select the layer you want. And the only tip I'll give you on this is just give them pins where you want them to move. So if we want her to have a knee, we need to make sure that we give her one. So kick her leg around, move her elbow. You'd be surprised--or maybe you wouldn't; you guys are probably all thinking how you'd use this-- how nice it is to have direct manipulation of an image in such a way. You can put those pins in there and just move it around. Really intuitive. Really, really easy to use. You can see I'm doing this all from quite a ways back on a trackpad. So that's what's going on there. Let's really briefly do a couple more things, and then I want to open it up to a few questions for you guys. Everything I've shown you so far is in CS5 Standard. Most of the Suites--I think all of the Suites--have Extended. The trial version has Extended, the educational version has Extended. Extended deserves its own sessions, and there are sessions specific to Extended. You guys should go talk to Zorana or listen to what she has to say because there is amazing stuff in there. But I want to give you just a really quick overview, and I'm going to use one particular file to do that. Extended has been around since CS3. That's when we first started doing two versions. The original one was a very ambitious release. It was architecture, engineering, measurements, medical formats, video, 3D-- all these features that built upon the regular release. And there was one thing that stood out for people, and they just jumped all over it, and it was 3D. They really wanted us to just run with 3D. They wanted to be able to paint in 3D and heal in 3D and stitch in 3D and wanted to bring 3D in and pass 3D out, and we did all that really quickly in CS4 Extended. CS5 Extended, the thing that we kept noticing is that creating certain 3D work is really difficult. You need to know 3D applications, and it can be really, really tough to extrude things. So say I've got a logo on a water bottle, a company logo, or maybe I want a type and I just want it to look different. If you want to do that stuff, you see it everywhere, but then you look into it and it requires knowing Maya and going really far with 3D and sort of having to have a 3D guy that you go to with all of that. And we thought, "We've got all this rich type, logos, vector." "We've got all these things in here. We've got a really rich 3D engine." "Let's see if we could present people with some presets "to do some of this stuff really powerfully and easily." That's a big, important part of this whole thing. So here we have standard vector shape of this logo here, and using presets that ship in CS5 Extended and this feature called Repousse, we can completely reimagine this. This could be a logo or it could be type, and we could see it as steel, we could see it shaded, extruded, lit, shadow. You have presets for all this, you can move it all around, and you can reinterpret it as you want. Just as easily plastic, inflated, softened, stamped plastic, aluminum, multiple light sources, text. You could do this stuff really quickly, really easily. Zorana, who is our Photoshop product manager for Extended, worked with a number of other people to do a book on 3D in Extended. There's a great iPad companion that goes along with it. She'll be doing a signing later today, I believe. Yes. 6:30 to 7:30 down at the 20th anniversary thing. I really recommend you check it out if you're interested in this. Really cool stuff, and there's a ton more than that, but that just gives you an idea. There's so much to it. Just to give you one more example of what can be done with that, you can actually take a photograph as well, which I will show you momentarily. And you could take a normal photo and you could select it and you could extrude that photo just like you're extruding type or anything else and you could light that and you could move that around and take 2D and turn it into 3D. But the thing you could do with Extended is you could take a 2D file and just make that what we call a 3D postcard, flip it around, do a reflection, model it onto a box or something. There's a tremendous amount of power there. Really, really interesting stuff. The other part that we won't even get into but it's certainly very topical is video. If you're playing around with video at all and you want to play around with it in a familiar place, Photoshop Extended can read video files. So you could make a black and white adjustment layer or do lens correction. All that stuff that I just showed you can be done to video as well. Okay. So we talked a bit about JDI earlier, just really brief background on how that came to be. The After Effects team did it. We knew it was a good idea. I went out to see the NAPP folks, the National Association of Photoshop Professionals, CS4. I did this huge demo for them. I did a five-hour demo, and I was really excited about it. I had actually just taken a 3D car, and I had spun it around and I had cut through the roof and I was painting the leather interior. I was feeling pretty good about my demo and kicked back and I said, "So, what did you guys think? It's pretty cool, huh?" And the first thing out of Scott Kelby's mouth was, "If you can do all that crazy stuff that you just did with that car and 3D and all that, "why can't you do this?" And then other people started chiming in, and we had another three-hour conversation about all of these little things. We had already been thinking about JDI, we had already been kind of wanting to do that. I left with a really good starting list for JDI. But I figured, "I've got this list of little things. Let's vet that with the public." I wrote a guest blog for Scott, I shared my email address with everybody. I've been doing that for a while. I really encouraged people to let me know what they would change. I asked our sales folks, our testers, our researchers, people in the field, other evangelists. With Photoshop we're really lucky we've got a really enthusiastic user base. And an interesting thing happened. We started hearing a lot of things. Oh. We also put Scott's blog on Facebook, which at that point we had about a half million fans. I had forgotten my email address was in there, so suddenly half a million people have my email address, and I got about another 1,200 emails. But a really cool thing happened. Even though we had a ton of requests, we realized a lot of them were the same requests and they were helping us prioritize the things we needed to do. It was really, really interesting. So what ended up happening? We thought, "Maybe we're going to do six or eight of these." I didn't know what we could do in a week because you've got to code it, test it, implement it, pens down at the end of the week. We did about three dozen of these, and I'll show you just a couple of them. Some of them are things like this, which if you're a new user and you want to straighten this image out, Photoshop feels a little bit like a cockpit of a 747. If you're a seasoned user, it's just a series of steps you need to go through. And so what you can do now is you can just take the Ruler tool and there is a Straighten button, and it rotates and crops that immediately for you. It just automates that and takes care of it right away. While we're here, let's say we have one of our Brush tools. We've got a nice little heads up display that shows you your color right here. You don't have to go and launch your color palette or open that up or click OK. You can just do that right onscreen here. You can get any color you want, and you actually have about six different configurations of that little widget right there. That's GPU accelerated. Let me show you another couple more. Let's say that we have an image that's-- We'll use this image for two things. Let's say that this is 16 bit, which it is. It makes it that much easier to say that. Someone gives me a RAW file, and I want to save that as a JPEG. If I'm a new user, I just can't save it as a JPEG, and that can be really confusing. I've seen people who deal with this. "Why can't I save it as a JPEG? I don't know what's going on." "I don't have that option anymore." They go crazy. They get very upset. If you're a seasoned user, you know Photoshop has half a dozen--[chuckles] 1,200-- secret handshakes. And one of them is you've got to switch that to an 8-bit image before you save it. So we figured, "Let's just take care of that for people." So if I go to save this 16-bit image, I now have JPEG as an option, not because 16-bit JPEGs have suddenly been invented but because we're doing that for you on the way through there. Again, it's just speeding up that workflow. So the other thing with this, let's take all the bubbles here. Let's take three layers. We've been able to multiselect layers for a while. But in this case what we want to do is we want to change the opacity of three layers. We have not been able to do that in the past. Imagine it's 300, and this would be that much more compelling. I can now change the opacity or the fill of multiple layers at once. So if you have 50 text layers and your client looks at something and they say, "It looks great, but I want you to adjust the opacity this much," instead of doing those one by one, you can now multiselect them and do that really quickly. People who deal with that sort of thing completely get the importance of that. The next one I'm going to show you--we'll make this the last one of these. This constitutes probably--I don't know--200 JDIs. I must have heard this again and again and again. Let's say I've got layer one here, and I want to put a drop shadow on that. I want that drop shadow to be a particular angle, distance, spread, size. I've got all the properties of that. I see people taking screen shots of these layer style dialogs so that they don't-- I'm sure some of you understand that; you're probably doing it--so they don't forget it. "This is the one that goes with this project," or, "This is the way I always do it." "I need it to be that way." And you have to change each and every one of these things every time you come in here. The stroke used to be red, then we changed it to black. I foolishly thought when we changed it to black that I was never going to hear anything more about that. But, "It should be blue," "It should be yellow." It should be the color you want it to be and the properties that you want it to be. So here's our garish stroke color. The good news is we can make any of these things our default now, so you guys can make any layer style your own default really quickly and easily. You can set it the way you want it and never have to deal with that again. So that's really, really easy to do that. Again, if you're in a production environment, it'll save you a ton of time. I promised you guys that I'd give you a few things here. Whole bunch of stuff--I don't know if you can see that, but I'll just sort of mention it. Adobe TV, I think we're going to have most if not all of these sessions there. We did that last time around. It's a great way to revisit these resources. There are trials, of course, on our page. NAPP has a dedicated page that's really helpful. John Nack does a wonderful blog that's got all sorts of things. Russell Brown has a great site. And I should mention social media, Facebook especially, Facebook and Twitter. I'm one of the people on the task force. We decide the content that goes on there. There's a lot of great information, there's some unique information. It's where we leak some features, it's where we ask people questions about things. So I really encourage you to join there and take a look if you haven't. And then last there at the bottom is my email address. I'm going to open things up to two minutes of questions, but I want to thank you guys for being here with me. I really appreciate it. [applause] Thanks. I don't know how much time we-- Two minutes. Do we have any questions? I'll repeat them so people can hear on the video. Are you going to save all your questions for email? [inaudible question] The question is I mentioned multiselect, but what about layers that aren't right there together like in different groups? Can I select those? Yeah, some of that did change around the CS3 time frame. We can talk a bit more afterwards, and I can give you some workarounds because there are some ways around old ways of doing things. Other questions? [inaudible question] So the question is around GPU acceleration. We do take advantage of the GPU. We're able to both accelerate our workflow and do some really incredible things but also present information onscreen that we wouldn't be able to otherwise. And we really don't have very stringent limitations on that. In Premiere's case, they are 64-bit native only, they've got a fixed set of cards, and they do absolutely incredible things with those cards. With us it's quite a bit more conservative than that. In fact, you can run on the lowest MacBook Air, you can run on the NVIDIA Ion, a $300 machine. If a Netbook supports the resolution, chances are not only are we running but we're running GPU enabled as well. We've found that more and more as time goes by if you're running a modern OS, your GPU is probably more than good enough to take advantage of our features set. So that's the short answer. I haven't seen a lot of problems with that at all. In fact, that's gotten a lot better over time. [inaudible question] I'm going to leave you guys with something. If you haven't said anything nice about me, this will be my opportunity to twist your arm. Okay. Let's say you're on CS3. Or let's say you're on CS5 because you're here listening to me talk about CS5. You have a friend who is on CS3. You both have a 5D Mark II, and you can open your files just fine, but she can't. Real bummer. She's going to have to upgrade. She spent all her money on her camera. Well, little known secret: If she were to pull down the free downloadable drag and drop DNG converter, which not only will future-proof her file but it will save 10 to 15 percent of her hard drive space, she can sneak that file into her old version of Photoshop. So there is a wonderful workaround. DNG is nothing but a good thing. If anybody wants to really hear me rant and ramble, that's a great thing to talk about. But you can actually take new file formats, new cameras, and bring them into old versions of our software. And in doing that, it encourages you to go with DNG, which that's not something we make money off of. Again, there are over 300 proprietary RAW formats, and we see that being a problem, so we proposed a solution that makes them smaller and more flexible. So that's the workaround. We're out of time. Feel free to come up and talk to me. Thanks again, you guys. I really appreciate it. [applause]